Stars
See inside any GGUF: where two quantizations part ways, token by token, and what each sampler cuts. Live in your browser, powered by llama.cpp (CUDA or Vulkan).
Doom E1M1's floor plan, played on a RISC-V CPU compiled into Qwen3.5-4B's weights (built on Qwenputer)
kyuz0 / ds4
Forked from antirez/ds4DeepSeek 4 Flash local inference engine for Metal and CUDA
Rust libraries and tools to interact with the Framework Computer systems
Blacksea is an active honeypot and canary-bait control system built to detect and drown LLM-driven attackers. Blacksea doesn't stop at watching LLM attacks. It exploits flaws in the attacker's LLM …
GLM-5.3-Flash EXL3 on 2x DGX Spark with TensorFold
The easiest, full-featured wireless router setup for Debian-based devices. Period.
The open-source AI engine for AMD Strix Halo. Text, embeddings, speech, image and video models in one binary, with OpenAI-compatible APIs, caching for coding agents and model extensions. Linux and …
Get the main content of any page as Markdown.
Skills for Real Engineers. Straight from my .agents directory.
Refusal removal as a loadable KV-cache graft — zero weight modification, no refusal-direction projection, fully reversible.
🕊️ A crowd-sourced guide to help techs help their non-tech spouses / partners / parents / kids when we are at the end-of-life
experimental linux kernel modules and userspace drivers for RDMA-over-usb4/thunderbolt on consumer hardware
ggml speech-to-text inference for 16+ model families
Режим идиоматического русского мата для AI-агентов. Короче, душевнее, эффективнее. 18+
A tool to unlobotomize your NVIDIA card!
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
cut Claude Code token usage by rendering text context as images
Port of Nvidia LocateAnything-3B on ggml
A QR code generator in a TrueType font: https://qr.jim.sh/
DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm
A FUSE-based virtual file system driver for dynamic OpenCode model discovery and configuration mapping.
Interactive BC-250 CU/WGP live manager using UMR, with TUI controls, safety checks, and boot-table persistence.
Re-enable all 40 CUs on the AMD BC-250 (gfx1013 / Cyan Skillfish). Kernel patch + build script. 1.61x compute scaling verified.
Super Productivity is an advanced todo list app with integrated Timeboxing and time tracking capabilities. It also comes with integrations for Jira, GitLab, GitHub and Open Project.